@paulag, In my opinion this is really effective **Analysis** and it's vital too because in this way we can understand, how many new users are entering into the **Steem Economy** and staying for long term and this understanding can gain by the **Retention** data. And i strongly believe that in future we will going to see more exciting opportunities in **Steem Economy** specially when we talk about the **SMT's** and then the **User Activity** will going to touch the moon height.

Wishing you an great day and stay blessed. 🙂

Great work @paulag . I always use your stats to check how ell the steem community is doing.

Keep dropping awesome content

Hey @paulag - Thanks for the wonderful analysis. From my experience with businesses, as you said, churn has always been a fact of life. There is no way to fulfill the needs of all of the users in one simple platform. 

I'm not sure how to access the data within STEEM, but maybe what we can start to do is to do cohort analyses. See if recent cohorts have drop-offs in activity at a much more accelerated pace. If not, I would say STEEM is here to stay! Also, it would be pretty cool to confirm whether or not price really does have an impact on churn (my logic says it is, but I don't have the concrete evidence to support this). I always say that this is the time to take *advantage* of the opportunity!

According to @themarkymark  (link below), there is between 8,000 and 12,000 active, real users.

From a project of over two years  online - I find it hard to see the positive evidence  in the 'progress' of steemit....(as the social platform).

10, 000 real users in 2 years !!!!

https://steemit.com/utopian-io/@themarkymark/did-hf20-reduce-or-increase-usage

Great Article - I agree that there are too many Bots on this platform. Humans are coming but it is a slow process. The next wave will make this platform very strong - the trigger?? - People can actually make money from blogging here...What a fantastic Medium!!

Great summary and analysis!  I think that as the price and market overall start to stabilize we will see improvements in churn.  I think it will also signal an improvement in overall engagement as those who stay provide more value to the ecosystem.  While still early, I think the HF20 impact will have a “knee-jerk” reaction where churn will increase but subsequently decrease as communities start to onboard their own users and create innovative ways to retain them and keep them engaged through RC delegation pools and/or SMT being developed.

Thank you very much for this information @paulag... And also thank you also for everything you do to try and help this community build and grow!  I know everyone's not always as hopeful and optimistic, so its nice to see you get the facts out there and then give us your opinions too.   Have a wonderful day :)

I talk to anybody willing to come off the beaten path, too,...

Can you look at how the value of votes increased, and how long it took to come back to normal, due to the three days of suppressed voting?
My napkin tells me i got a 50% bump in my vote's value.
I know that isnt because 20k redfish stopped voting.

Before @statsmonkey earned the ire of the un named one his report said 10 voters were distributing ~30% of the daily wealth.
If bid bots are distributing ~30%, too,  how much is left after the top 100 vote?

Im sure the inflation dilutes this sad fact, but shining a light on the curent facts can only help, imo.

> Churn = Number of Active users at the beginning of the period + New registrations – Number of Active users at the end of the period.
>
> Churn rate can be calculated using 
>
> Churn rate = Churn / Number of Active users at the beginning of the period.

Assuming that we are defining "active users" as accounts which have posted a comment or post during the period, and "new registrations" actually means "new registrations total during the period", I'm not sure that this function actually properly shows the churn rate.

Instead it should probably be:

ChurnRate = TotalNewRegistrationsOverPeriod / TotalActiveUsersOverPeriod

Logically, that would mean that as the number of active users increases for a set number of registrations, churn rate decreases, and as the number of new registrations in the period increases for a set amount of activity, churn rate increases.

(And ultimately, I'm not sure that this is a good definition of churn in the first place. What we really want to know is specifically, from period to period, how many accounts which are created never become active, and I'm pretty sure that we have done that particular lens before.)

> *STEEM is not only used a as a social networking type blockchain.  Some people are investors and wallet holders only.  Previous analysis have shown this makes up less than 0.5% of accounts on STEEM and so no adjustment has been made as I have deemed it immaterial.

Is this really still the case? I feel like this sort of statement either should link to an article where that analysis is made, or we have to start by defining our terms and actually deriving the real numbers.

Would an easy way to differentiate these kinds of accounts be to simply look for accounts which have never made a post or a comment, but which are carrying at minimum 30 SP? That feels like the right breakpoint to eliminate accounts which were simply created and never made active, the very definition of churn.

That's not to say that I don't think it might not be less than 0.5% of accounts which qualify there, but it's not something that we can take upfront granted. (And given the account creation system post hardfork 20, I suspect we'll see some change in that breakdown.) 

> The churn rate varies considerably from week to week as it is dependent on two variables. The distinct users and The number of new accounts.  As both of these variables vary from week to week it is expected to see this variation reflected in the churn rates. 

The churn rate by your calculations swings wildly from -30% to over 30%. Likely as a side effect of the fact that you are calculating Churn as a difference. Churn rate should never be negative, and that should have been your first indicator that your method wasn't showing what you expected.

You can have a lot of churn or you can have a little churn, but you can't have negative churn.

Again, this should have been an indicator that something wrong is broken.

> Personally I feel the current churn rates are perfect.  If we can maintain these levels and grow the accounts we have the STEEM blockchain and SMT's will have an improved chance of success.  The next question is, how can we accelerate the growth of accounts while we have this lull to take advantage of?

I would take some exception to that, looking at the actual rate of content being produced and where it is being focused, it's pretty clear that we are seeing less general interest content being forged – and specifically that is why people come and bring value to the steem blockchain. Outside of the social network side of the operation, it really doesn't have much to offer someone who is just "into cryptocurrency." And as long as the content being created for the social networking side of the steem blockchain is monocultural, cryptocurrency for crypto cultists, we will never, ever see actual value being brought into the blockchain because there are far, far better places for investors to park their funds.

If the vision is that the steem blockchain is and should essentially be a blockchain ecology of digital applications intended purely solely to talk about the steem blockchain itself, then we are on a fine track. If it's to be a general-purpose social network ecology that just happens to be backed on a blockchain and gets certain advantages there from – then a lot of things have to change.

But first we have to start with accurate churn numbers. Once we have usable numbers, we can make better assessments.

Thank you for all the analysis ✅ It will be really interesting to see the influence of HF 20 in another one or two month... Hoping the best 😊  @peekbit

Hi @paulag, very interesting follow-up to the Q2 report! While the new registration counts continue to decrease over time, the churn rate seems to follow this trend - which is not too bad if I understood it correctly?! I must admit, I was also a bit puzzled if the churn rate can be negative. Given your definition, it makes sense if previously inactive accounts are reactivated. Do you have insights how the churn rate is calculated for the other social networks? They must have the same effects somehow. But what I'm unsure about: does it make sense to add new accounts unconditionally to the churn or should only those be included that are actually "active"? I've seen definitions of the churn rate as `(active_users_at_the_beginning - active_users_at_the_end) / active_users_at_the_beginning`. I think this could treat inactive new accounts more accurately?

Your contribution has been evaluated according to [Utopian policies and guidelines](https://join.utopian.io/guidelines), as well as a predefined set of questions pertaining to the category.

To view those questions and the relevant answers related to your post, [click here](https://review.utopian.io/result/0/1222313).

---- 
Need help? Write a ticket on https://support.utopian.io/. 
Chat with us on [Discord](https://discord.gg/uTyJkNm). 
[[utopian-moderator]](https://join.utopian.io/)

#### Hi @paulag!

Your post was upvoted by @steem-ua, new Steem dApp, using UserAuthority for algorithmic post curation!
Your post is eligible for our upvote, thanks to our collaboration with @utopian-io!
**Feel free to join our [@steem-ua Discord server](https://discord.gg/KpBNYGz)**

Thanks.  Nice to see some positive news on the subject.  I at least feel that we have a good grasp of the problems now even if we don't as a community have all the solutions yet.  

Hey, @paulag.

I'm guess I'm okay with the conclusion that the churn rate is falling, but I'm wondering about the factors involved. Before I get to that, though, I should say, even if it is what I think it is, I'm okay with it to a degree. I think it's a good thing for those of us who are still here. Maybe not so much for the ecosystem.

I guess if we're looking at the churn rate in percentages based on the numbers that were previously coming/leaving versus what they've been up to hard fork 20, I've got to wonder if it's come down because a) there were fewer accounts coming in, so b) they were able to find a community quicker and c) the larger amounts of folks who left because of the decline in STEEM prices basically since April-ish (if memory serves) was the worst of the exodus or churn. Meaning, that those who have stayed have found other reasons for it other than earnings.

I have no data to back that up other than looking at what you've provided. I'm wondering if we haven't just reached that core group of humans (plus bots, apparently) that are going to hang on as long as STEEM is here, while trying to bring the newbies into the fold.

The missed observation from all of this is how when you consider the number of total unique individuals on Steemit, then the churn rate over time, and calculate how many people are staying on the platform vs # of the same people creating new accounts (bots, alts, etc), we've lost more people than we've been gaining and the ones that remain are simply exploiting the rewards pool for personal gain by creating more new accounts to grow their nestegg.

Hey, @paulag!

**Thanks for contributing on Utopian**.
We’re already looking forward to your next contribution!

**Get higher incentives and support Utopian.io!**
 Simply set @utopian.pay as a 5% (or higher) payout beneficiary on your contribution post (via [SteemPlus](https://chrome.google.com/webstore/detail/steemplus/mjbkjgcplmaneajhcbegoffkedeankaj?hl=en) or [Steeditor](https://steeditor.app)).

**Want to chat? Join us on Discord https://discord.gg/h52nFrV.**

<a href='https://steemconnect.com/sign/account-witness-vote?witness=utopian-io&approve=1'>Vote for Utopian Witness!</a>